Blocking Kappa-Opioid Receptors Extends Ketamine’s Antidepressant Effects in Adult Mice Stressed as Teens

The impact of chronic unpredictable stress (CUS) and pharmacological treatments on locomotor activity and depression-like behaviors was assessed using established behavioral paradigms: the open field test (OFT) [27] and the tail suspension test (TST) [28]. No significant differences were detected in the OFT parameters—total path length and center path—between the control and CUS groups or among the treatment groups at either 24 h or one week post-treatment (Figure 1B).

Depressive-like behavior was characterized by the TST at 24 h and one week post-treatment. TST was selected as the main readout because it is a widely validated, highly sensitive paradigm for detecting the effects of rapid-acting antidepressants like ketamine in C57BL/6J mice [28,29]. Given the large number of animals and the requirements of the experimental design, TST was selected as the most appropriate, as it allows rapid and reliable assessment of behavioral despair and is well-suited for high-throughput experimental design [28,30]. CUS exposure significantly increased immobility time in stressed animals at both 24 h (mean difference 98.3; Cohen’s d = 3.13; 95% CI 48.40 to 137.9) and one week (mean difference 59.75; Cohen’s d = 2.24; 95% CI 22.97 to 96.53) compared to controls (Figure 1C).

To evaluate the effects of CUS and pharmacological interventions—including individual (ketamine or nBNI) and combined (ketamine/nBNI) treatments—on locomotor activity and depression-like behaviors, behavioral tests (BTs) were conducted 24 h and one week post-treatment time points using the open field test (OFT) and the tail suspension test (TST). The experimental timeline is illustrated in Figure 1A.

Twenty-four hours post-treatment, all interventions significantly reduced immobility compared with the CUS group (ketamine: mean difference = −68.92; Cohen’s d = 2.22; 95% CI 27.21 to 110.6; nBNI: mean difference = −94.38; Cohen’s d = 2.21; 95% CI 52.67 to 110.6; Ket/nBNI: mean difference = −83.64; Cohen’s d = 1.91; 95% CI 44.46 to 122.8), confirming their short-term antidepressant efficacy (Figure 1C, left). However, at one week post-treatment, neither individual treatment maintained the antidepressant effect (Figure 1B). In contrast, the combined Ket/nBNI treatment produced a sustained reduction in immobility, effectively restoring it to control levels (Ket/nBNI: mean difference = −60.25; Cohen’s d = 2.63; 95% CI 26.20 to 94.30), demonstrating a prolonged antidepressant effect (Figure 1C, right).

To delineate the molecular substrates associated with the behavioral effects of chronic stress and pharmacological intervention, we examined key intracellular signaling pathways governing neuroplasticity and stress adaptation within the prefrontal cortex (PFC), hippocampus (HIPP), and striatum (STR). Western blot analyses were employed to quantify the activation state of ERK, AKT, JNK, and mTOR kinases, conveyed as phosphorylated-to-total protein ratios. Together with BDNF signaling, these proteins collectively regulate synaptic plasticity, neuronal resilience, and affective homeostasis [31,32].

To further characterize signaling relationships associated with the combined Ket/nBNI treatment, correlation analyses were performed between upstream kinases and mTOR activation across brain regions (Figure 5). In both PFC and STR, JNK activation was inversely correlated with mTOR activation (r = −0.8308 and r = −0.7785, respectively) (Figure 5A,B), indicating that lower JNK activity was associated with higher mTOR activation in these regions. A similar inverse association was observed in the PFC between ERK and mTOR activation (r = −0.8279) (Figure 5C). Although these correlations do not establish direct cause-and-effect interactions, they suggest that the activity of these pathways changes in parallel under the combined treatment condition. Notably, the consistent negative relationship between JNK and mTOR across cortical and striatal regions represents a recurring molecular pattern linked with the treatment-associated signaling profile.

In the PFC, CUS significantly reduced BDNF expression compared to the control group (mean difference = −83.75, 95% CI [−106.6, −60.88], Cohen’s d = 8.3) (Figure 6A). None of the treatments reversed the stress-induced alterations, indicating persistent impairment of neurotrophic signaling. In contrast, in the hippocampus, CUS significantly elevated BDNF expression relative to controls (mean difference = 65.02, 95% CI [30.41, 99.64], Cohen’s d = 8.3) (Figure 6B), suggesting a possible compensatory response to chronic stress. Surprisingly, the combined Ket/nBNI treatment did not have any effect on CUS-induced BDNF increase in the hippocampus, although ketamine and nBNI treatments were able to attenuate this increase (ketamine: mean difference = 50.72, 95% CI [16.11, 85.34], Cohen’s d = 2.34; nBNI: mean difference = 39.23, 95% CI [2.05, 76.4], Cohen’s d = 1.79). In the striatum, CUS profoundly suppressed BDNF expression (mean difference = −70.33, 95% CI [−90.62, −50.03], Cohen’s d = 5.3) (Figure 6C). Although all treatments significantly increased BDNF levels, the highest increase was achieved with Ket/nBNI and nBNI treatments (nBNI: mean difference = −66.45, 95% CI [−86.74, 46.15], Cohen’s d = 4.55; Ket/nBNI: mean difference = −55.01, 95% CI [−75.30, −34.71, Cohen’s d = 5.5) that fully counteracted the stress-induced BDNF suppression. Ketamine treatment only partially reversed this suppression, increasing BDNF levels (mean difference = −25.27, 95% CI [−45.56, 4.97], Cohen’s d = 1.91) (Figure 6C).

In order to explore the relationship between BDNF expression and behavioral performance across the examined brain structures (PFC, HIPP, and STR), correlation analyses were performed (Figure 6D–F). BDNF expression showed significant correlations with immobility time across all three brain structures, although the direction of these correlations differed between regions. In the PFC and STR, a significant negative correlation was observed (r = −0.45; r = −0.50, respectively) (Figure 6D,F), whereas in the hippocampus, the correlation was positive (r = 0.46) (Figure 6E).

The long-term consequences of adolescent CUS were evident in adulthood as persistent depressive-like behavior accompanied by robust, region-specific alterations in stress-related intracellular signaling pathways (Table 1). In the prefrontal cortex (PFC), adolescent stress was associated with reduced trophic support and disrupted kinase balance, consistent with impaired top-down regulation of affective and stress responses. ERK activation was reduced, consistent with chronic stress impairing trophic and survival signaling [33], whereas AKT signaling remained unchanged. Paradoxically, increased mTOR activation, despite reduced upstream kinase activity, may reflect long-term maladaptive circuit remodeling following stress [34]. Alternatively, it could be linked to stress-induced metabolic dysregulation and increased body weight, as chronic mTOR hyperactivation is known to promote anabolic processes and contribute to obesity [35]. JNK activity was markedly suppressed in the PFC, suggesting long-term downregulation of stress-responsive kinase activity following chronic stress, as was reported previously [36,37]. The long-term reduction in JNK activity may be associated with a rebound increase in mTOR signaling, but this shift occurred without restoration of upstream trophic support. Consistent with this, BDNF levels in the PFC remained significantly decreased long after CUS and showed a negative correlation with immobility time, suggesting that persistent deficits in neurotrophic signaling may contribute to the maintenance of depressive-like behavior [38].

In contrast to other regions, the hippocampus displayed a molecular profile consistent with long-term compensatory or metaplastic adaptation rather than a persistent acute stress state. Although chronic stress often reduces hippocampal BDNF [39,40,41], its direction and magnitude can vary with stress timing and duration, sometimes resulting in recovery or even upregulation [42]. In our study, elevated BDNF alongside preserved baseline mTOR activity suggests a region-specific adaptation emerging during prolonged recovery from adolescent stress.

Meanwhile, the striatum showed alterations in stress- and trophic-related signaling compatible with modified reward-circuit regulation. The striatum displayed a distinct stress-response pattern characterized by elevated JNK activation and reduced AKT signaling, with ERK and mTOR levels unchanged. This pattern may reflect altered stress sensitivity or signaling imbalance, potentially biasing the region toward pro-apoptotic or maladaptive signaling [43]. Finally, BDNF expression in the adult striatum was significantly reduced long-term following chronic stress, as reported previously [44,45].

The present findings of opposite-direction BDNF–behavior correlations across brain regions are consistent with a growing body of evidence indicating that stress-induced BDNF modulation is highly region-dependent and can participate in distinct adaptive or maladaptive responses depending on local circuit demands and stress history [46]. Mechanistically, BDNF levels are also highly dynamic and influenced by interactions with glucocorticoid signaling and other modulatory systems. Reviews of stress-related neuroplasticity illustrate that BDNF expression not only varies across brain structures but also fluctuates over the course of acute and chronic stress exposure and that these variations emerge in parallel with changes in glucocorticoid receptor signaling and downstream intracellular cascades that shape synaptic remodeling [47]. Under this framework, elevated hippocampal BDNF in stressed animals may reflect a compensatory or metaplastic process, especially following prolonged stress or recovery periods, while suppressed BDNF in prefrontal and striatal regions may index persistent trophic deficits related to stress pathology. Such region-specific BDNF dynamics are consistent with the notion that stress does not uniformly dampen or elevate neurotrophic signaling but instead reprograms plasticity responses according to the unique functional and receptor profiles of each circuit, thereby shaping both behavioral vulnerability and resilience [42].

Together, these findings support the idea that early-life stress leaves a long-lasting heterogeneous molecular imprint across corticolimbic and striatal networks, which may shape both vulnerability to depression and responsiveness to later interventions.

The sustained antidepressant-like effect observed after combined Ket/nBNI treatment was accompanied by region-specific modulation of intracellular pathways involved in stress adaptation and synaptic plasticity, including ERK, AKT, JNK, mTOR, and BDNF. The concurrent NMDA receptor and KOR blockade was associated with structure-dependent kinase changes converging on mTOR and BDNF regulation, rather than with a uniform molecular response.

Among the pathways examined, JNK—a stress-responsive kinase and downstream component of KOR-related signaling—was of particular interest. Given that stress has been reported to regulate JNK in a region-dependent manner [48,49], concurrent modulation of NMDA receptors and KOR may be associated with differential patterns of JNK-related signaling across corticolimbic and striatal circuits. In addition, prior work indicates that KOR blockade can be accompanied by alterations in components of the NMDA–BDNF–mTOR signaling cascade under stress conditions [50], suggesting that combined glutamatergic and KOR modulation may coincide with molecular states consistent with plasticity-related processes.

We deliberately selected a single 6 mg/kg dose of ketamine [51] for the following reasons: to balance antidepressant-like efficacy with minimization of nonspecific behavioral effects (e.g., hyperlocomotion or psychotomimetic-like activity) that can occur at higher or repeated doses [52]. Limiting the study to a single administration and defined post-treatment assessments reduced confounds related to repeated injections, cumulative drug exposure, and handling stress, thereby allowing clearer evaluation of the pharmacological interaction between ketamine and KOR antagonism. The one-week post-treatment time point was selected to assess behavioral persistence beyond acute pharmacological effects, as ketamine and its metabolites are cleared well before this interval [53,54]. Although nBNI has prolonged receptor-level activity [25], nBNI alone did not produce sustained behavioral effects at one week [55], supporting the interpretation that the combined treatment reflects longer-lasting functional adaptations rather than residual drug presence. Importantly, nBNI is not solely a neutral receptor antagonist but has been described as a biased ligand capable of recruiting JNK and producing prolonged functional desensitization of KOR [56,57]. Such prolonged receptor modulation has been associated with changes in neurotrophic signaling pathways, including mTOR and BDNF [29,58]. In the present context, these signaling alterations are interpreted as molecular correlates observed alongside sustained behavioral effects rather than as evidence of direct mechanistic mediation.

The correlation patterns observed after the combined Ket/nBNI treatment provide additional context for interpreting the region-specific signaling adaptations. In both the PFC and striatum, the inverse association between JNK and mTOR activation indicates that lower JNK activity coincided with greater mTOR pathway engagement under the treatment condition. Although correlational analyses do not establish regulatory directionality, the recurrence of this relationship across two anatomically and functionally distinct regions suggests a coordinated shift in signaling balance, whereby reduced stress-associated kinase activity accompanies enhanced plasticity-related pathway activation. A similar inverse relationship between ERK and mTOR in the PFC further demonstrates that upstream kinase activity does not uniformly scale with mTOR output, underscoring the complexity of intracellular signal integration within stress-responsive circuits.

Importantly, these associations should not be interpreted as evidence of direct inhibitory interactions between pathways but rather as coordinated signaling states associated with the combined pharmacological intervention. ERK and mTOR are both downstream effectors of glutamatergic and trophic signaling and can influence each other through intermediary regulators. In addition, strong mTORC1 activation is known to engage negative feedback mechanisms involving S6K1 and IRS proteins, which can reshape upstream AKT and MAPK signaling [61]. The inverse ERK–mTOR relationship observed in the PFC may therefore reflect network-level feedback and pathway redistribution rather than a linear regulatory interaction.

Collectively, these data propose a model in which Ket/nBNI treatment engages region-specific plasticity mechanisms. While the PFC and hippocampus show signs of normalized signaling, the striatum exhibits a unique profile of mTOR and BDNF upregulation that parallels the sustained antidepressant-like effects. While direct functional validation is needed, we posit that this distinct upregulation of striatal plasticity may contribute to the extended therapeutic duration observed in the combined treatment group.

Therefore, the persistence of behavioral improvements observed with the combined Ket/nBNI treatment cannot be attributed solely to ketamine or interpreted as strict pharmacological synergy. Rather, the findings should be interpreted as a combinatorial association in which sustained receptor-level modulation by nBNI may create a permissive molecular environment that supports the prolonged engagement of plasticity-related pathways by a single ketamine dose.

Although our study provides insights into the complex molecular consequences of chronic unpredictable stress (CUS) and the therapeutic potential of ketamine and KOR antagonism, several limitations should be considered when interpreting the findings.

First, experiments were conducted exclusively in male mice, which limits conclusions regarding sex-dependent behavioral and molecular responses. However, it is important to emphasize that the results remain internally valid and biologically meaningful within this experimental context, as male rodents are widely used and well-characterized in preclinical stress and ketamine research. Substantial evidence indicates that females may exhibit distinct—and in some cases divergent—responses to ketamine and stress-related manipulations. For example, ketamine has been reported to produce sustained antidepressant-like effects in males while inducing delayed anxiety- or depressive-like behaviors in females [62]. Female rodents may also respond to lower ketamine doses, with ovarian hormones influencing treatment sensitivity across the estrous cycle [63,64]. Accordingly, inclusion of both sexes in future studies will be essential to determine the generalizability of the present findings rather than to establish their validity [6].

Second, the study employed a single ketamine dose and a single defined post-treatment time point. This design limits inference regarding repeated dosing regimens, dose–response relationships, and longer-term trajectories beyond one week. Clinically, ketamine is often administered repeatedly, and whether sustained KOR antagonism would extend durability under such paradigms remains to be determined. Nevertheless, isolating a single administration allowed us to specifically test the durability of an acute pharmacological interaction without confounds related to tolerance, cumulative neuroplastic adaptations, or repeated handling stress. This focused design strengthens interpretability regarding persistence beyond acute exposure.

Third, molecular analyses were performed at a single post-treatment time point, which limits conclusions regarding the temporal evolution of the observed signaling changes. Nevertheless, the selected time point was sufficient to capture stable behavioral effects and associated molecular signatures, providing a meaningful snapshot of stress- and treatment-related adaptations. Future studies incorporating longitudinal molecular profiling will help clarify the persistence and progression of these changes.

Fourth, the behavioral assessment in this study relied primarily on the tail suspension test (TST). While TST is a robust, widely validated, and highly sensitive paradigm for detecting the effects of rapid-acting antidepressants like ketamine in a large number of experimental animals, we acknowledge that it focuses on immobility as a measure of behavioral despair. The inclusion of additional behavioral paradigms, such as the sucrose preference test to assess anhedonia or the forced swim test, would provide a more comprehensive behavioral profile in future investigations. In addition, repeated behavioral testing may act as a stressor, potentially influencing molecular signaling [65]. To control for this, all groups—including controls—underwent the same testing schedule, ensuring that observed molecular differences reflect treatment and stress effects rather than unequal test exposure. Therefore, molecular findings should be interpreted as reflecting both treatment/stress and the impact of repeated testing.

Finally, although coordinated molecular changes were observed across multiple brain regions following combined ketamine and KOR antagonism, the present study was not designed to directly test mechanistic causality. Potential contributions of receptor-level interactions, intracellular signaling convergence, or circuit- and cell-type-specific processes therefore remain to be examined. Importantly, the region-specific signaling patterns identified here provide a focused and testable framework for future hypothesis-driven mechanistic studies.

Despite these limitations, the findings in this study provide novel preclinical evidence that KOR antagonism can prolong the antidepressant-like effects of ketamine and identify region-specific molecular adaptations associated with sustained behavioral improvement.

All experiments were conducted on adolescent male C57BL/6J mice acquired from the Jackson Laboratory and bred in our facilities. Animals were provided with commercial food pellets and water ad libitum and maintained on a 12 h light/dark cycle in a temperature-controlled environment (20 ± 2 °C). The animals were housed in groups of three to four per cage. All animal procedures were approved by the Ethical Committee of the VINCA Institute of Nuclear Sciences (323-07-03268/023-05), in accordance with the guidelines set by the European Communities Council Directive (2010/63/EU) for animal experiments.

During the experiment, adolescent animals from different litters were randomly divided into two main groups: a control group and a group exposed to chronic unpredictable stress (CUS). An investigator not involved in behavioral testing or molecular analyses randomly assigned animals to the control and CUS groups, making sure that animals in each group were from different litters and from different cages. A similar principle of randomization was applied for the allocation of CUS animals to treatment groups (as described in). Section 4.3

Chronic stress is a major contributing factor in the development of mood disorders such as depression and anxiety, often leading to persistent behavioral and neurobiological alterations. To explore these effects, we employed the chronic unpredictable stress (CUS) paradigm, a well-established and widely validated model for inducing depression-like phenotypes in rodents [6,7,66].

Adolescent male mice were exposed to CUS for 12 consecutive days (Table 3), starting on postnatal day 28 (P28). The 12-day CUS protocol was selected based on prior validation in adolescent mice [6,7,66], a developmental period characterized by heightened stress sensitivity, during which shorter stress paradigms are sufficient to induce robust and persistent depressive-like behavioral phenotypes. Stressors were administered three times daily during the light phase (morning, afternoon, and evening) in designated procedure rooms. A variety of physical and social stressors were used each day to prevent habituation and ensure sustained stress responses. Between stressors, animals were returned to their home cages. Following the CUS procedure, mice were given a 30-day rest period (P40–P70) to allow maturation into adulthood before further testing and treatment.

Animals not exposed to the CUS were used as controls (n = 10). To ensure consistency across all animals, the control group received an acute mass-adjusted volume of vehicle (0.9% NaCl) by intraperitoneal (i.p.) injection.

Drug administration: Animals exposed to the CUS were randomly divided into 4 groups (n = 10 per group) for the i.p. drug administration on P70: I CUS group was treated with the mass-adjusted volume of vehicle (saline, 0.9% NaCl); II Ket group was treated with ketamine (Ketamidor, Richter Pharma AG, Wels, Austria; dosage of 6 mg/kg) diluted in saline; III nBNI group was i.p. injected with nBNI (Sigma-Aldrich, St. Louis, MO, USA; dosage of 10 mg/kg) diluted in saline [56]; IV Ket/nBNI group was i.p. injected first with ketamine (6 mg/kg) and then with nBNI (10 mg/kg) diluted in saline, 30 min apart. A dose of 6 mg/kg ketamine was chosen based on previous studies demonstrating effective antidepressant-like effects in mice at this dose while maintaining a subanesthetic profile [51].

To evaluate the effects of CUS and pharmacological interventions—including individual (ketamine or nBNI) and combined (ketamine/nBNI) treatments—on locomotor activity and depression-like behaviors, behavioral tests (BTs) were conducted 24 h and one week post-treatment using the open field test (OFT) and the tail suspension test (TST). After the last BT was performed at one week post-treatment, the animals rested for 24 h and were then sacrificed, and brain structures (prefrontal cortex, hippocampus, and striatum) were isolated and immediately frozen at −80 °C. The experimental design is presented in Figure 1A. Molecular analyses (n = 8) were conducted 24 h after the last TST (which itself was conducted one week post-treatment) in order to minimize acute behavioral-test–induced signaling effects. In addition, all control animals used in the study underwent the same testing regimen.

Body weight (g) was measured at three time points: at the beginning of adolescence and immediately before the onset of the CUS protocol (P28), at the end of the 12-day CUS exposure (P40), and after the rest period when mice reached adulthood (P70). Monitoring of body weight served as an indicator of general health and as a physiological marker of chronic stress exposure. Measurements were performed individually using a precision digital scale at consistent times of day to minimize variability. Reduced body weight is a well-established physiological indicator of chronic stress in rodents [67].

CUS-exposed adolescent mice (P40) exhibited a significant reduction in body weight compared with control animals (p < 0.0001, Student t-test) (Figure 7), confirming the efficacy of the chronic stress procedure. However, by P70, CUS mice showed a marked increase in body weight relative to control animals (p < 0.001), indicating the greater weight gain during the rest period.

The frozen brain tissues (n = 8) were then weighed and homogenized (1:10, w/v) in ice-cold 50 mM Tris-HCl (pH 7.5) buffer containing 150 mM NaCl, 1% NP-40, 0.1% SDS, 10 mM EDTA (pH 8), 10 mM EGTA (pH 7.2), 0.5% Triton x-100, and phosphatase inhibitors (20 mM b-glycerophosphate, 5 mM Na₄P₂O₇ × 10 H₂O, 2 mM Na₃VO₄, 25 mM NaF) with 30 strokes of a Potter Elvehjem Teflon-glass homogenizer. Each sample was centrifuged for 30 min at 14,000 rpm at 4 °C, in order to obtain the supernatant. The samples were subsequently frozen and stored at −80 °C prior to performing Western blot analysis.

Protein concentration was determined using the method of Markwell [68], and samples were incubated for 5 min at 100 °C in an appropriate amount of denaturing buffer, as described by Laemmli [69]. Forty μg of protein was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 7.5%, 10%, or 12% gels. The proteins were then transferred to a polyvinylidene fluoride (PVDF) membrane (Immobilon-P, Millipore, Darmstadt, Germany) using a Transblot transfer system (Bio-Rad, Hercules, CA, USA). Following transfer, the membrane was blocked with 5% non-fat milk in phosphate-buffered saline (PBS) for 1 h at room temperature. The membrane was then incubated overnight at 4 °C with primary antibodies diluted in 2.5% non-fat milk in PBS-Tween (PBS-T), as listed in Table 4. After overnight incubation with the primary antibodies, the membrane was washed three times with PBS-T. After incubation, membranes were washed three times with PBS-T and then incubated with HRP-conjugated secondary antibodies (rabbit anti-mouse IgG-HRP and goat anti-rabbit IgG-HRP; Abcam, Cambridge, UK). Immunoreactive bands were visualized using enhanced chemiluminescence (ECL) with SuperSignal Pico or Femto substrates (Thermo Scientific, Waltham, MA, USA) and exposed to X-ray film (Amersham Hyperfilm ECL, Cytiva, Wilmington, DE, USA).

Chemiluminescent signals from the immunoblots were detected and quantified using ImageJ analysis software (1.53k, NIH, Bethesda, MD, USA) for densitometry of protein bands on X-ray film. After quantification, protein levels were normalized to Ponceau S-stained total protein bands (40–70 kDa) as a loading control. To enable comparison across different gels, an internal reference standard (IRS)—an identical sample loaded on every gel—was used. For Western blot analyses, samples were coded prior to processing and quantification, and the investigator performing densitometric analyses was blinded to group identity. Relative protein expression in each sample was calculated as a percentage of IRS, ensuring consistency and allowing quantitative comparison between gels.

Data were analyzed using the Prism program (GraphPad Prism, Software, v.6, La Jolla, CA, USA). Data distributions were assessed for normality (by the Shapiro–Wilk test) and for homogeneity of variance (by the Brown–Forsythe test) prior to analysis. Sample sizes for the study were determined using a priori power analysis based on our previous experimental evidence of CUS behavioral effects (power = 80% at α = 0.05). The comparison between the two experimental groups was performed by an unpaired Student’s t-test. For comparing multiple groups, one-way analyses of variance (ANOVA) were performed, followed by Dunnett’s post hoc test to determine statistical significance between the CUS group and other experimental groups. For molecular analyses, 95% confidence intervals for mean differences and effect sizes (Cohen’s d) were reported as indicators of the precision and magnitude of the observed effects. According to Cohen’s guidelines [70], Cohen’s d values greater than 0.8 were interpreted as indicators of a large effect. The outliers were excluded using the criteria median ± 2 MAD [71], and additional exclusions were made in cases of technical issues or sample loss. Pearson correlation coefficients (r) were calculated to assess the associations between expression levels of different proteins and between immobility time and protein expression levels. All data are expressed as mean ± standard error of the mean (SEM). The test was considered significant when p < 0.05.